There's a moment that happens to every operator eventually. A founder who's been grinding for two years messages you at 11pm. The book is bleeding, the team is burnt out, the contracts are rolling off. They want out. You can have it for pennies — sometimes literally a dollar plus assuming the contracts.
The temptation is to say yes immediately. The reality is that most underperforming agencies are underperforming for structural reasons, and you can absorb the same structural problem just by signing the paper.
Seven days is enough to figure out whether the book is salvageable. Here's the audit.
Day 1: The cash and contract reality
Before anything else, the math has to work. If the cash math doesn't work, no operational fix matters.
Pull the contract list. For each retainer, you need: monthly fee, contract end date, auto-renew clause, services promised, and payment status. The number you actually care about is net retention if 30% churn out in the next 60 days. That's the worst-case scenario you're inheriting, and it's almost always closer to reality than the founder's optimistic projection.
Then pull the AR aging. Anything more than 60 days late is functionally written off — assume it. If 25% of the AR is in that bucket, the book is worth 25% less than the founder is showing you.
If the math survives those two filters with at least 30% margin to your operating costs, keep auditing. If not, walk now and offer to consult instead of acquire.
Day 2: The infrastructure inventory
Most failing agencies fail at the infrastructure layer. You need to know what you're inheriting before you commit to running it.
Walk through every tool that has a recurring charge: GoHighLevel sub-accounts, email tools, voice AI minutes, Clay credits, Calendly, Stripe, Slack, project management. Note the renewal date and whether the founder owns the account or a contractor does. Contractors holding root credentials is the single most common landmine in agency acquisitions.
Then look at the integration topology. How many of those tools are wired together? How many are duplicating each other? An agency that pays for two CRMs, three email tools, and a phone system that doesn't talk to any of them is paying $4-7K/month in tool costs that an operating system approach consolidates to under $1K.
The good news: integration debt is fixable. The bad news: it usually takes 3-4 weeks of operator time to clean up, and the founder will hand you a list of "things we'll fix soon" that has been on the list for a year.
Day 3: The client health audit
For every client on the book, score them on three axes: payment health (do they pay on time?), engagement health (do they respond to emails, attend QBRs?), and outcome health (are they getting results from the service?).
A client that scores green on all three is gold — protect them obsessively in the transition. A client that scores red on payment but green on outcome is salvageable with a contract restructure. A client that scores red on outcome regardless of the other two is going to churn within 90 days; price the acquisition assuming they're already gone.
The hardest conversation in week 1 is identifying clients you'd actually rather lose. There are usually 1-3 of them on every underperforming book — clients who are paying but consuming 5x their fair share of operator time, draining team morale, and producing churn risk on the rest of the book through delivery delays. A planned, graceful exit for those clients in week 2-3 is one of the highest-leverage moves you can make.
Day 4: The team diligence
If the agency has any contractors or employees, the team audit happens day 4. Three questions per person:
- What do they actually do — measured in deliverables, not hours?
- What would happen if they left tomorrow?
- How replaceable is what they do with the right AI agent plus a 10-hour-a-week human?
Most underperforming agencies have one critical person doing 60% of the load (often a delivery operator) and 2-3 people whose work could be absorbed by automation. The acquisition math changes a lot depending on which bucket each person falls into.
Be honest with yourself about which roles can be automated. There's a real case here — see the agent classification framework for which roles are actually AI-replaceable in 2026 and which still need a human in the loop.
Day 5: The acquisition channel audit
How is the agency getting new clients today? Trace the last 10 client signings to source. Cold outreach? Referrals? Inbound from a podcast or content? Paid?
The reason this matters: the acquisition channel often dies in the transition. A founder who closed via personal network can't transfer that channel to a new operator. A founder who closed via cold outreach has a system you can take over. A founder who closed via paid ads has spend you'll have to keep paying or lose the pipeline overnight.
If 70%+ of the client base came from a channel that doesn't transfer (founder's personal network, founder's content, founder's podcast appearances), you're not buying an agency — you're buying a client list with a 12-month decay curve. Price accordingly.
Day 6: The product-market fit signal
Look at the cohort of clients who joined in the last 6 months versus the cohort from 12-18 months ago. Compare retention. If the recent cohort is churning faster than the older one, the offer has stopped working — usually because the niche has moved, the comp set has improved, or the agency's edge has eroded.
This is the most consequential signal in the audit. An agency with strong recent-cohort retention is a fixable infrastructure problem. An agency with collapsing recent-cohort retention is a positioning problem, and positioning problems take 60-120 days to fix even when you know exactly what to do.
Day 7: The decision
By day 7, you have enough to answer one question: in 90 days, can you make this book healthier than it is today, with the cash and bandwidth available?
The decision usually splits three ways:
- Acquire and stabilize if the math works, the team is salvageable, and the recent-cohort retention is intact
- Acquire as a wind-down at a steep discount if the cash is fine but the channel doesn't transfer; you milk it for 12 months while building your own
- Walk if the recent-cohort retention is collapsing or the AR is more than 30% rotten
There's no shame in walking. The operators who succeed at agency rollups are the ones who walk from 8 of 10 deals.
Where AcquireOS fits in the audit
If you're considering a takeover, the easiest way to model the post-acquisition operating cost is to look at what the consolidated stack would cost on a single platform. The Operator tier handles a single-niche book of 8-15 retainers with one delivery person. The Agency tier handles a multi-niche book at 25-40 clients. Your audit math should compare the founder's current tooling spend to what the same client base costs on platform — the delta is usually 60-70% in your favor and pays for the integration time within the first 60 days.
The principle of week-one diligence: every assumption gets verified, and the cash math is the floor. Everything else is operational improvement.
